PROJECT SUMMARY There is a critical gap in our knowledge pertaining to the key psychometric properties of the measurement tools we use to assess chronic pain in severe cerebral palsy (CP), especially in the context of repeated assessments and performance in response to clinical procedures. If precision medicine is ever to become a reality for those with cognitive, communicative and motor impairments associated with severe forms of CP (hereafter `severe CP'), there is a great deal of foundational work yet to be done. The long-term goal is to create a reliable `pain phenotype' model that is clinically relevant to making treatment decisions for patients with CP. The overall objective in this application is to establish a set of assessments (i.e., pain assessment battery) that are feasible for use with patients with severe CP, and that demonstrate strong psychometric properties. The central psychometric hypothesis is that pain assessment scores will be adequately stable over time, will vary together, and will significantly decrease following pain treatment. This hypothesis has been formulated based on preliminary data collected by the investigative team. The rationale is that testing the score properties related to consistency (reliability) and sensitivity to change of the pain assessment battery are essential to defining pain endpoints. The study hypothesis will be tested by completing three specific aims: 1) determine whether pain measures, repeatedly sampled, produce psychometrically sound scores; 2) determine whether the pain measures are sensitive to change in response to putative pain treatment; and 3) identify data elements in Aim 1 that predict pain variability assessed in Aim 2. To accomplish the aims, pain will be repeatedly assessed prior to and again following corticosteroid joint injection to reduce chronic joint pain. Assessments include caregiver report of pain and psychosocial factors, blind observationally coded pain behaviors scored during a standardized pain examination procedure, and a modified quantitative somatosensory test. The first two aims will establish temporal and internal consistency, and convergent and construct validity. Aim 3 will be completed by conducting regression analyses to identify data elements that may predict treatment variability. The proposed research is innovative because it leverages complex statistical modeling to simultaneously capture the multidimensional aspects of pain (i.e., intensity, interference, duration), provides item-level parameter calibrations and latent trait estimates, results in multidimensional factor structures, and allows for meaningful detection of intra-individual change. New research possibilities are expected to become available as a result. The proposed contribution is the documentation of key psychometric properties of the pain assessment battery for use in severe CP. This contribution will be significant because it will position the field with dependable pain measures with which to further explore treatment efficacy and individual differences in treatment response. Ultimately, such knowledge will lead to development of a pain phenotype for CP and support an individualized medicine approach.